Abstract
Background: In a time when the incidence of severe traumatic brain injury (TBI) is increasing in low- to middle-income countries (LMICs), it is important to understand the behavior of predictive variables in an LMIC's population. There are few previous attempts to generate prediction models for TBI outcomes from local data in LMICs. Our study aim is to design and compare a series of predictive models for mortality on a new cohort in TBI patients in Brazil using Machine Learning.Methods: A prospective registry was set in São Paulo, Brazil, enrolling all patients with a diagnosis of TBI that require admission to the intensive care unit. We evaluated the following predictors: gender, age, pupil reactivity at admission, Glasgow Coma Scale (GCS), presence of hypoxia and hypotension, computed tomography findings, trauma severity score, and laboratory results.Results: Overall mortality at 14 days was 22.8%. Models had a high prediction performance, with the best prediction for overall mortality achieved through Naive Bayes (area under the curve = 0.906). The most significant predictors were the GCS at admission and prehospital GCS, age, and pupil reaction. When predicting the length of stay at the intensive care unit, the Conditional Inference Tree model had the best performance (root mean square error = 1.011), with the most important variable across all models being the GCS at scene.Conclusions: Models for early mortality and hospital length of stay using Machine Learning can achieve high performance when based on registry data even in LMICs. These models have the potential to inform treatment decisions and counsel family members.Level of evidence: This observational study provides a level IV evidence on prognosis after TBI.
Highlights
In a time when the incidence of severe traumatic brain injury (TBI) is increasing in low- to middle-income countries (LMICs), it is important to understand the behavior of predictive variables in an LMIC’s population
Models had a high prediction performance, with the best prediction for overall mortality achieved through Naive Bayes
Models for early mortality and hospital length of stay using Machine Learning can achieve high performance when based on registry data even in LMICs
Summary
In a time when the incidence of severe traumatic brain injury (TBI) is increasing in low- to middle-income countries (LMICs), it is important to understand the behavior of predictive variables in an LMIC’s population. As one of its primary examples, the Glasgow Coma Scale (GCS) has been widely used to predict outcomes, associating patients with a score of 13–14 with longer post-traumatic amnesia and a higher rate of abnormal brain image findings at 6 months after the initial trauma. The GCS has the potential to present its predictive performance enhanced by the inclusion of other variables, as has been demonstrated in the improvement of prediction accuracy of hospital mortality [6] through the integration of variables such as age and brain stem reflexes [7] In another example, when the GCS was combined with the Injury Severity Score, their joint performance significantly improved in comparison with isolated scores or the Abbreviated Injury Score for outcomes measured 12 months after the initial injury [8]. Prediction models including multiple variables tend to exceed isolated, manually calculated scores [10], opening up an opportunity for the use of prediction models as they can increase predictive performance [11, 12]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
